Getting out of a Rut

I realized recently I was starting to get a little bored by about my 4th or 5th battery of a session. I’m used to that happening over the winter when I’m stuck flying in-doors, but I’ve not had that happen when I’m able to be outside before. I’m guessing it’s just because I’ve just been flying my yard for the last couple of months since the pandemic started, whereas I’m used to going to a nearby fields and schools to get a change of scenery. I probably could still do those things, but it just seems a like an unnecessary risk to take no matter how insignificant.

Thinking on it more, I’ve been subconsciously feeling this for a few weeks now and just hadn’t realized it until now. I’d started flying my whoop outside along with my toothpick, and I think that was my way of trying to mix things up a bit.

Besides just flying the same small space over and over again, I’ve also still been in that winter/spring mindset where I feel like I have to get out and fly whenever the weather is decent, and there’s been a lot more of that recently as our weather here has started to shift more into summer than spring. So I’ve been pushing myself to fly as much as possible when the weather is good and feeling guilty when I don’t, instead of resetting my expectations for the bad weather days being the exception rather than the norm.

The biggest issue though has been how I fly. I don’t have any kind of goals or plan or structure to my sessions right now. Last summer and fall, I was all about practicing tricks and trying to get more comfortable in the air. This year I started out doing some of that by trying to get more comfortable doing loops and lookbacks where I have to be upside down, but recently I’ve just been doing mostly the same comfortable things repeatedly with some minor changes like starting to do more low to the ground proximity flying.

To correct that and get out of the rut I’ve gotten myself into, I’ve tried to become more intentional with my flight sessions. I’ll take a battery to warm up where I just do my normal messing around, and then for the next couple of batteries I’ll pick a maneuver or trick that I want to work on and just do that over and over for most of the pack.

For example, I’ve noticed recently that I tend to turn left most of the time. Probably 98% of the time that I do a dive I go left. Same for yaw spins. For rolls I’ve noticed I most often go right for this and for flips I pitch backward. All of these has to do with the stick movements. I fly what’s known as Mode 2, which most pilots do in the US, and that puts the yaw and throttle on the left stick and pitch and roll on the right stick. So yawing left means pulling the stick to the outside of the controller, as does rolling right.

Somehow I’ve gotten more used pulling the sticks than pushing them towards the center. So that’s a big part of my practice now is forcing myself to reverse the way I do tricks and the flight lines I pick to get more used to yawing right, rolling left, and pitching forward. It’s almost a different kind of boredom, zipping back and forth across my backyard doing yaw spins one way five or six times and then doing them the other direction. But in other ways it feels good, as I can feel how uncomfortable I am when doing things differently and I definitely don’t want to be.

Tree Bonk

Wednesday last week we had some good weather and I was able to wrap up work by 5 pm, so I charged up six batteries and headed out to the back deck to fly. I was having a pretty good session, doing a little bit of racing flying different circuits around the house and yard, doing some freestyle as well, just generally having fun and not really trying to work on anything in particular.

I started really having fun on to run my loops around the backyard that I used to do with my whoops, and found doing it with a bigger quad is pretty exhilarating. It’s a lot faster and the space feels tighter even though my toothpick quad isn’t that much bigger. I was getting pretty good too at finding different lines to follow. Trying for smoother lines each time around while mixing in a little freestyle here and there as well. It was a lot of fun.

Then I bonked the tree in the front yard. I thought it was pretty minor at first since I was just banking to do a 180 around the tree and go back the other way. It’s something I’ve done hundreds of times with no problem, except this time I pulled it up too short and hit the tree trunk. Not very hard, but I wasn’t able to recover it, so I had to do a walk of shame out front to get it. Which is when I discovered the broken prop on the front right motor, and that explained why I couldn’t recover and take off.

No big deal, I replaced the broken prop with a new one and tried to take off and finish the rest of the current battery. But when I armed it I only had three props spinning. The front left motor wasn’t working.

I disarmed the quad and spun that motor by hand, but it wouldn’t turn freely. It felt like something was catching, and I figured either the bell was bent or the magnets or stator might have gotten damaged inside. I was pretty surprised I had motor damage though, I didn’t think I’d hit that tree hard enough for that.

So that was the end of my session for the day. Later that evening I took the prop off and got the motor pulled out of the quad in preparation for replacing it. Out of curiosity I took the tiny clip off of the bottom of the motor so I could remove the bell and see what the actual damage was. That was when I discovered that there wasn’t any damage just a tiny grain of dirt or sand stuck in between two windings on the stator which must have been sticking out just enough to catch on the magnets inside the bell. I pulled that dirt out with a pair of tweezers and put the bell back on to test it and the thing spun just as smoothly as when it was new.

While it was unexpected and great that I didn’t need to replace the motor, I’d lost the clip while getting it apart. At the time I didn’t care because I wasn’t expecting to be able to put it back on the quad, but now I needed that tiny clip to keep the motor together. There was no way I was finding a tiny black piece of metal like that in my carpeted office though. Luckily I’ve been keeping all my broken parts including motors in a box, and I was able to get the clip off of one of them and use that one the motor.

This was a first for me. I normally end up just replacing parts, I’ve never been able to fix something and put it back on. Once I got the motor back on, I plugged a battery into the quad and tested the motor out and it worked great.

As a side note, the 2″ tri-blade props I’d switched to once I ran out of the 65mm bi-blades have proven to be a little too brittle. I broke too many clockwise props and didn’t have any more replacements, so I dropped back to some 1.9″ ones that I’d bought an never used. I’m going to need to order some new props though at this rate.

Here’s the full flight just to prove I don’t just crash all of the time.

Stupid Mistakes

It was a nice day and I was flying from my backyard while the kids’ were playing on their trampoline. I’d just started my fourth pack and was flying around the front of the house when my nose started to itch. What I should have done was landed to take care of it, but what I actually did was try fly one-handed and take care of it. I believe the words “real quick” passed through my mind. I ended up with me drifting into the side of the house, bouncing off, and crashing.

I knew I was in for some problems though because right after that the feed to my goggles went blue. I’ve never had that happen before. That can only be bad though.

Static usually means that the VTX is dead. A black screen with the on screen display means there’s a problem with the camera. But a blue screen was new to me, so I wasn’t sure what the problem was.

I tried unplugging the battery and plugging it back in, but got the same blue screen in my goggles. Taking off the canopy, I didn’t see any damage to the wires from the flight controller to the video transmitter or from the VTX to the camera. I tried the battery again with the canopy off and didn’t see any difference. Taking the VTX out of the canopy didn’t help either. I accidentally notice that if I squeezed the VTX between my thumb and index finger the picture would pop back in briefly. So I started looking more closely at the board and finally figured out the problem.

The red arrow is pointing to a tiny brown and silver rectangle. That’s a capacitor. Here’s a closer photo.

Another one with tweezers included for scale.

So the crash had knocked that capacitor partly loose. Whatever it’s purpose it was allowing the VTX to transmit enough that my goggles were picking up a signal but not actually sending any OSD or the camera image. It was also far, far too small for me to try to resolder. I did try to put some liquid tape over the area, thinking that might hold it in place, but it was so small that just putting that on the board moved the capacitor.

I’ve really had a hard time with this A01 VTX. More so than anything else that came on the HX100 originally. But I’m not sure I can blame this one anything other the me being stupid and having to bad luck to crash into a landscaping timber. Regardless of the cause though I was ready to try something else.

I’d never been a fan of the soldered on antenna connection for one, especially since it came with a dipole antenna and I used left-hand circular polarized antennas on my goggles and my whoops. So this was a change to upgrade the VTX antenna as well as the VTX itself.

I’d been pretty happy with the Eachine Nano that I’d used in my brushless whoop build, and had bought a second one as a spare, so I decided to use that as a replacement. The trick was figuring out how I was going to mount it in the quad since it’s rectangular shaped and wouldn’t mount in the canopy in the same way. I also didn’t want to try to stick it to the flight controller with double-sided tape. After a bit of fiddling around trying to decide how to mount it I had the idea to strip all of the components off of the old A01 VTX board, mount it in the canopy, and then stick the new VTX to it.

I had to rewire things a bit since the new VTX didn’t have a camera passthrough on it like the original one. So I could run the 5V and ground from the VTX to the camera, but had to splice in a new wire from the camera to the flight controller (the yellow to blue signal wire below).

Once the new wiring was figured out, I just had to solder everything back onto the flight controller, which by this point is pretty routine for me. Plugged in a battery to test that I could see a picture in my goggles and that I was able to change the VTX band, channel, and power from the OSD. Then I put heat shrink on the VTX to both isolate it electrically from the old VTX I was going to mount it to and also to hold the antenna on. This board uses a UF.L connector which is very light but isn’t the most secure, so I like to have the heat shrink on it to help make sure it doesn’t go anywhere.

That did require an update to my canopy mod though, since the new antenna wouldn’t fit through the old hole that I’d drilled. So I clipped the end just a bit to allow the antenna to slip through. I debated drilling a hole through the end that I could zip-tie shut, but so far it’s seemed pretty secure without it.

So now I have a further modified HX100. I think at this point maybe one or two motors, the canopy, and the camera are the only original parts. Most of the problems I’ve had with it have been due to crashes and my own stupid mistakes. It’s definitely been my favorite quadcopter to fly.

Still Flying

Despite having a couple of projects in mind, I haven’t been doing much over the last two weeks except flying. With everyone stuck at home, I’ve just been sticking to my yard which is kind of limiting, but after being stuck flying whoops indoors for the winter having decent weather now to get outside regularly is a treat even if I can’t get to a bigger field or go over to the nearby elementary school’s parking lot.

I started practicing proximity flying a lot more, which just means flying close to objects and trying to hit smaller gaps. I have a loop in my backyard that I like to do with whoops that I started trying with my toothpick and while I’ve been getting better at it, I have had a couple of collisions. My favorite so far was then I was trying to skim our fence line and caught a board. I couldn’t manage this again if I tried.

While that collision didn’t do any damage, I whacked the bird feeder that we keep hanging from my kids’ playset that ended up breaking a motor. I’m honestly surprised about how much damage it did compared to some of the other collisions and crashed. It might be that was one of the older motors on the quad, since I don’t keep track of which was which when I’ve done rebuilds but that’s probably something I should be better about doing. It might have just been the speed I was going. Regardless, I broke the bell again and had to swap the motor out.

I recorded myself doing the motor repair and experimented to making a timelapse of it that I posted online. Normally I use YouTube for anything longer than what will fit on Twitter or Instagram as a clip, like this one of the 30 seconds before the crash.

But I hadn’t realized I could post videos of 10 minutes or less to Instagram’s TV service and it would show up in my regular feed. While I always try to post things to both Twitter and Instagram, I’ve noticed that on Twitter I normally get feedback from my friends and on Instagram I get more feedback from other FPV pilots. Which was why I was excited by the IGTV find because I can post YouTube links on Twitter but not on Instagram. So I was able to put the timelapse motor repair on both YouTube/Twitter and IGTV and it was interesting to see how different the audiences were on both platforms.

Approaching Twitter and Instagram from the point of view of a content creator has been very new to me. I’ve always used both platforms to just share things and interact with friends that I already have, but as I’ve been doing more quadcopter stuff I’ve been trying to use hashtags and things to try and reach a broader audience.

Besides getting a bit better at Instagram, I’ve done a couple of experiments with Twitch streaming. I did two rough test streams and then a longer one where I did about 10 batteries and posted online when I went live. I had a receiver that I could connect to my tablet via an USB cable and then capture and stream the quadcopter camera feed using a Streamlabs app. That all worked better than I expected from a technical standpoint. It even looks like I could add the front facing camera on the table to show me wearing the goggles when I’m flying, but I haven’t tried that yet. I’m not sure about Twitch as a platform for sharing FPV flying though, as I can’t interact with chat while I’m flying. I can try to catch up on messages in between batteries though, which would mean about 3-5 minutes of flying followed by chatting for a couple of minutes and then flying again.

It’s an interesting idea and something I’m going to try again soon, along with doing more videos for Instagram and YouTube.

Fly Hard, Break Stuff

The longer I fly quads, the more comfortable I get, and the more risks I start to take. As a result I end up crashing a fair bit, not as often as when I first started, but these crashes involve more speed and end up causing a lot more wear and tear on my quad. For example, here’s photos from 3/14 and 3/23 just over a week apart.

The props are so shiny and smooth with clean edges in the one on the left, whereas the one on the right has been crashed a few times and the props have whacked tree branches, pavement, and the gutter along the edge of my garage. The frame has taken some serious frontal hits and started to split a bit on the end of one of the arms. You can also see a scratch on the motor bell too.

I’ve found that I’m taking a fair bit of pride in how used it looks after about a week of flying. It means I’m pushing myself trying to get better. Taking risks. Crashing doesn’t scare me at this stage since I’m getting enough experience with repairs that I don’t think there’s anything I couldn’t replace in it. In fact, thinking about it now, I think there’s probably two motors that are original, plus the canopy, and the camera. Everything else the frame, the flight controller, the video transmitter, and the receiver are all upgraded or replacements.

All that said, I’ve been grounded the last few days because of the weather. We had a couple of days of rain, and I thought I was going to be able to fly on Sunday but I hadn’t realized that we were in for 30 MPH winds. That’s too much for me to handle for yard flying when I might catch a gust at the wrong time and hit my house or worse the neighbor’s. I was almost able to sneak in some time on Thursday last week but it literally started raining as I walked outside to set up, so I had to turn around and go back inside. Rather than just discharge all my batteries and give up through I used them to charge up a set of 1S batteries and flew my whoop inside. It’s not quite the same but it’s still fun.

1S Charging Adapter

When I first got into flying I was using a whoop style quad that flew on a 1S LiPo battery, which means a single celled lithium-ion polymer battery. The kit I bought came with a small USB charger that would handle up to four batteries and took quite a while to charge them completely.

I soon upgraded to a parallel board that would do six batteries and had a DC wall converter that would handle 3 amps, USB 2.0 caps out at 0.5 amps for comparison. It was much faster and better but it couldn’t do one important thing. After flying, I basically had to charge the batteries back to full, and Lipo batteries have a much better longevity if they’re left charged to what’s called storage voltage, basically about half way charged at 3.8 volts and about 30-50% capacity. Technically I could keep an eye on the charger while it was working and stop it once the batteries hit storage voltage, but sitting and watching batteries charge is not any more fun than watching paint dry.

Unfortunately there weren’t any chargers on the market at the time that would do storage charging for 1S batteries. I was looking at a charger geared for higher cell-count batteries anyway as I already knew I was going to be getting into bigger quads, and I happened across a forum page that talked about making an adapter cable to charge 1S batteries on multi-cell chargers. The adapter works by combining all of the 1S connectors in series to make the individual batteries look like cells in a single battery to the charger. It didn’t look too hard and would be a good way for me to practice soldering.

To make one I needed 1S female connectors, silicone wire, a balance lead, and an XT60 power connector. The number of 1S connectors depended on the size of the balance lead. I went for a 6S to be able to charge as many 1S batteries at a time as possible. This did mean that I couldn’t just charge one or two batteries on an adapter like this, I’d always have to do six at a time and all six needed to be the same capacity.

Soldering these up in series involved soldering the red wire on the yellow XT60 connector and the red wire on the balance lead to the red wire of the top-most 1S connector.

From there it was just a matter of connecting the next black wire on the balance lead to the black wire of the 1S connector along with the red wire of the neighboring adapter, and repeating across all 6 balance lead wires until getting the the last one where the black wire of the XT60 is included in the solder joint as well. It was an ideal project for someone getting into the hobby and needing practice soldering. Especially when I forgot to slid the heat shrink over some wires before soldering them, so I’d have to desolder the joint and redo it with the heat shrink in place.

The last stage was to add some heat shrink over the whole adapter. This last bit isn’t to electrically isolate any solder joints, but just to bind all of the wires together and make the adapter easier to use.

Balance chargers like my HOTA D6 Pro Dual use the combination of the power and balance leads to make sure that all of the individual cells in a battery are charging evenly. This let me treat six 1S 300mAh capacity batteries as a single 6S 300mAh capacity battery and more easily charge, storage charge, or even discharge batteries than any of the options I had at the time.

Just recently there’ve been parallel charging boards that have come out that will do 1S batteries by adding a jumper from a balance port to a power port. This loses one spot on the parallel board but it means that the charger treats the batteries as a single 1S battery with five times the capacity, so the charger can push more amps and charge the batteries faster. It also isn’t limited by needing exactly six batteries like my original cable so I can charge anywhere between one and five batteries, since the sixth slot is take up by the jumper.

Repairs Revisited

The replacement flight controller came in so I could follow up on the repairs from last time, where I realized I had two bad ESC’s after finishing the rebuild. When flashing my new replacement for the replacement flight controller, I noticed that there were two firmware targets that were very similar: MATEKF411 and MATEKF411RX. I had picked the RX version thinking that was for FC’s with built-in receivers. I’m not sure that’s what the RX means in this case as there’s also an SE target. It doesn’t matter anyway since these FC’s don’t have built-in receivers, I was confusing it with the flight controller that I had recently used on my whoop build.

So for the new flight controller I smartly used the correct firmware, the MATEKF411 version, and the new board tested out fine both in Betaflight and BLHeli. All the ESC’s were recognized and I even plugged in motors and made sure they all spun up. Then I had the thought that maybe the problem with my other FC was because of the wrong firmware. So I plugged that in and flashed the same firmware on that, and checked in BLHeli and all four ESCs were recognized. The board and ESCs were fine all along, I just had the wrong firmware loaded so the resource mapping was incorrect. I’m honestly surprised that it worked at all but I’m guessing that the major difference between the two models is the pin assignments on the main chip. I’m glad I didn’t brick it or burn anything out.

I honestly wasn’t expecting to prove out my comment from last week so soon, but there really is always something new to learn in this hobby.

Now I had two working flight controllers. One was all soldered up and just needed to be remounted on the carbon fiber frame, I even had some extra foam to use since I’d torn off the foam padding that was originally stuck on figuring I wouldn’t need it. The other FC was better designed for a toothpick style frame with it’s sideways mounted USB and motor plugs, but it would have to be soldered to the receiver and video transmitter. So I had to decide on which way to go. While I really liked the design on the new board, I decided to use the other one since it was already soldered and just needed to be mounted. I’d save the new flight controller for when I busted the current one or decided to build a new quadcopter.

I remounted the flight controller and other electronics on the frame, and got the quad rebuilt in time for a little test flight one evening when there was decent weather. The first battery went well. For the second one I bumped the video transmitter power up from 25mW to 200mW, which gives me a better picture when I have obstacles between me and the quad. After making that change the video started to go to snow and then cut out whenever I would throttle up. I was not happy. I’d only got one flight in before a new problem had popped up.

So the next two nights I spent a lot of time troubleshooting the problem. I checked solder joints. I looked for damage to wires and electronics. I tried with the canopy off so the VTX was hanging loose from the flight controller, and noticed that it worked better that way, so I tried a different set of screws and standoffs to have the canopy sit higher off of the flight controller but that didn’t work either. I tried replacing the wires on the VTX harness. I tried direct soldering the wires from the flight controller to the video transmitter. Nothing worked and eventually I’d run out of ideas to try. The problem seemed to be getting worse too as it was a now having issues when the VTX was loose, and started to cut out completely until I restarted it. So I gave up and swapped out the video transmitter for a new one. I’d bought two the last time I needed to replace it, so I already had a spare.

After getting the new video transmitter installed everything worked perfectly. I’m not sure what caused the last one to fail. Possibly I messed up something when soldering, either the board got too hot or I bumped something. I don’t think so, since I normally notice those things. It might just have been cumulative damaged from crashes and just vibration form flying that finally stopped it from working. There’s no way for me to know, and it’s one of the hardest things for me to deal with in this hobby, not being able to determine exactly what happened and why.

We had cold but sunny weather on Sunday, so I was able to get get out and fly my yard for a full six packs before my fingers got too cold. I didn’t have any more trouble with the quad. Finally.

I’ve been very happy so far with the all of the flight controllers, canopy, motors, and frames that I’ve bought from BetaFPV but I think I’m done with their A01 VTX’s. I’m on my third one for a quad that I’ve had for about seven months, and it seem like they should last longer than that. So I’m looking at some other manufacturers for alternatives to switch to when this one dies. Or maybe I’ll find that a couple of months is just about as long as these last. Or I’ll get a surprise and this one will end up lasting for a year.

Repairs and Frustrations

Sunday was a great day for flying, especially for March. It was sunny, there wasn’t much winder, and it hit the mid 60’s. Very rate for this time of year. So when I got up in the morning I made it a priority to get my toothpick repaired, as it was still sitting in pieces from the prior weekend’s crash at the local elementary school.

I’m starting to thing that place is cursed for me, as only a couple of months ago I whacked a pole and broke a motor. On my second battery I caught a branch and crashed to the pavement, which snapped on of the zip ties that I use to mount my receiver antennas, and I found later when I opened it up also knocked the smartaudio wire off of the flight controller.

So I sat down at my hobby table and got set up to do some soldering. I had decided in addition to just fixing the wire and the antenna I was going to finally get around to troubleshooting why my smartaudio wasn’t working.

Smartaudio is a protocol that allows the flight controller to change the band, channel, and power on the VTX. For some transmitters this is a nice bonus and saves a pilot from having to read tiny LEDs and do a bunch of button presses to switch settings. For my VTX thought it’s the only way to change the settings, since it doesn’t have a button and the LEDs only show what the power level is. Up until now I’d just been setting the goggles to the band and channel the VTX got stuck on, but if I ever wanted to go somewhere with other pilots I wouldn’t be able to fly like that. I also wanted to be able to increase the transmitter power, since the VTX would go up to 200mW and provide a much clearer picture when going around obstacles.

I resoldered the wire and tested out the smartaudio but it still didn’t work, which I expected. I’d previously checked the wire and the solder joints, so I didn’t really think resoldering the wire again would fix things, but I wanted to check regardless. The next step was to swap the VTX. I’d been putting this off because I didn’t want to have to move the camera over to the other VTX that I had since it would require desoldering and then soldering three wires. Not that I’d mind doing that, but I’d rather not when I don’t know if the VTX is really the issue. I had gotten a VTX and camera combo though for a possible replacement part for my brushless whoop, if and when it was needed, and it used the same plug and pin-out that my toothpick VTX did. So I was easily able to plug in the new VTX and camera and see if the smartaudio worked there. It didn’t, which was good as it meant I knew the problem was in the flight controller, but bad since that meant I’d have to swap in a new flight controller. Which basically meant a rebuild.

As I was taking the quad apart so I could pull the flight controller out, I found a crack in the front of the frame. I’m not sure if it was from that last crash or not, but it turns out that there was a silver lining for having to swap the flight controller. I probably wouldn’t have noticed that crack otherwise, since I tend to focus on the arms and motor mounts when looking for damage.

Just the foam that the flight controller was mounted to was holding the front together. I found the start of a crack at the rear mounting screw as well, and there was enough stress on the frame when I pulled the mounting foam off that the frame split in half.

Once I had everything disassembled, I soldered up the new flight controller to my receiver and my VTX and tested out the smartaudio again and found it working. So it was some kind of problem with the original flight controller. Not too surprising I suppose given the amount of abuse it’s taken over the last six months or so.

At this stage I was pretty pleased with myself, I’d finally resolve and issue that had been bugging me for months. I got out a new frame and started rebuilding my quad again. I took this as an opportunity to adjust one other thing that had been bothering me.

When I originally upgraded the receiver, I put it on top of the flight controller. Originally it was underneath and I didn’t like that since I was worried about the battery impacting it in a crash, so I put it on top. The problem with that is the canopy is a pretty tight space and it was tough to get the wires for the VTX and the receiver along with the VTX and receiver themselves all tucked in under the canopy without crimping any wires. So this time around I fixed the flight controller mount so there would be a bit more space underneath and moved the receiver back there. I also added more battery padds around the area thinking that would help protect the receiver and the USB port.

My last step in the build was to hook the quad back up to the computer and check the motor directions. I didn’t keep track of which motor was in which position when I took them off of the old frame, so I figured some of them would be backwards and I’d need to change some setting in BLHeli.

I plugged in the quad and when into Betaflight to the motors tab and tried to spin up the first motor, only to have nothing happen. Motor two spun up but it was in the wrong position, I was setting the motor two throttle in Betaflight but it was the motor in position three that was spinning. Motor three spun up but was in the motor four position, and motor four didn’t spin at all. So basically the right-side motors weren’t spinning and the left-side motors were, but they were mapped to the wrong motor positions. Remapping motors isn’t hard, it’s just a couple of commands that need to be entered in Betaflight’s command line interface tab:

resource MOTOR 1 B07
resource MOTOR 2 B06
resource MOTOR 3 B05
resource MOTOR 4 B04 

The problem was I had two motors not spinning up at all. I switched over to the BLHeli configuration tool and it would only read two of the ESCs, 2 and 3 of course.

So now I’m stuck. I’ve no idea if I did something to the flight controller when I was soldering or mounting it to the new frame, or if I just got a bad one with two dead ESCs on it. I’m guessing they were just bad as all of that circuitry is on the bottom of the board and I did all of my soldering on top. But either way it meant I wasn’t able to go out and fly. I mean I could have rebuilt it again with the old flight controller but then my smartaudio issue would be back, and that would be undoing a lot of work.

In the end I went online and ordered a new flight controller. I took the opportunity to get one that was more designed for toothpick frames, it has the USB and motor plugs side mounted instead of sticking straight down. That should make them harder to damage, and will be easier to maintain.

I’ve made a point in my whoop build log and this one that you should test out your flight controller right out of the package, and I guess I need to add another step to that. For these flight controllers with the battery leads already soldered on and motor plugs, it’s just as easy to add motors and plug in a battery and test that all of the ESCs work as well.

I love this hobby. Flying is a mixture of exhilaration and relaxation for me. Soldering electronics and building a quad is very satisfying. There’s always something new to learn in this hobby though, and it can be very frustrating to put a couple of hours into troubleshooting a build only to be stuck with an unflyable quad while waiting on new parts.

Brushless Whoop Build – Part Two

Last time I left off at having finished the soldering and successfully tested that I could connect with my radio to the flight controller and receive a video signal on my goggles, and that the video signal was right side up. I tested that the VTX protocol was configured correctly so I could use my radio to bring up the VTX settings in the OSD on my goggles and change the VTX band and channel, and that the frequency the VTX was transmitting at changed appropriately.

Everything worked as expected, so aside from the one solder bridge issue that I had to fix the first part of the build was pretty trouble free. At this point I figured that the hard parts were done as the rest of the work was just assembling the pieces onto the frame, which normally isn’t any harder than building a Lego set.

First step is to put the rubber mounting grommets on the corners of the flight controller and then mount it in the whoop frame. The grommets are hollow and fit over plastic posts that have threaded screw holes in them. I put the screw in at the rear of the flight controller as that one wouldn’t be covered by the camera mount, and it left me get the flight controller partially secured into the frame.

The next step was to get the camera snapped into its mount and then fit that over the two side and front grommets of the frame, and get all three remaining screws in and snugged down. One of the major bad habits that I’ve had to overcome with building quads is over tightening screws. Especially in instances like this where doing so flattens out the grommets and cancels out any vibration protection that they might provide for the flight controller. Or worse stripping out the plastic posts so that the screws don’t hold at all.

Once I got the camera mounted, it was time to get the VTX on. I’d tried a couple different options before starting the build and I thought I’d figured the best way to get the VTX and antenna mounted. I started by cutting some clear heat shrink to keep the VTX from touching the back of the camera and shorting out. It also had the added benefit of holding the antenna in place so I didn’t have to worry about trying to secure that as well. With the heat shrink on then I zip tied it upright to the back of the camera mount with the status LED and button visible.

With all of that out of the way, all that was left was the motors and props. After doing the first motor, I discovered it was easier to plug them into the flight controller before screwing them into the frame. So that’s how I did motors two through four.

Putting the props on requires some attention to detail since props come in two versions clockwise and counter-clockwise, and they have to be put on in a certain order. For my setup, I prefer what’s known as reversed props or props-out. Which means that the front right and rear left props spin clockwise, and the other two counter-clockwise. There’s some aerodynamic benefits to setting up the props this way, Oscar Liang has a good article on that.

Usually I don’t have any problems figuring which prop is which and getting them all on the correct motors, but I ran into a novel problem with the new Azi props I was using. I got one of the motors put on upside down and didn’t realize it until all four props were on. I’d never had a problem with putting a prop on upside down before. To make things worse, these were push fit and didn’t need or use screws. Which meant they were extra hard to get back off in order to fix them.

In the end because the whoop frame was pretty flexible and the props were on so tightly, I ended up having to unmount the motors so I could get a good enough grip on them to pry off the props and get everything back on right-side up and in the correct order.

The last step was to hook the quad back up to my computer and launch the BLHeli utility to make sure that all four motors were spinning the correct direction. In this instance motors two and three were backwards, but it’s an easy software change to make and write to the ESC’s. Once that was done the build was finished.

All together the new whoop weighs 24 grams and 32 grams with a 350 mAh battery, that’s just 2 grams more than my brushed Acrobee.

The last step naturally is to fly it. I charged a couple of batteries up and plugged the first one in to do the maiden flight and ran into an issue where only three of the motors spun up after the quad was armed. I tried disarming and then arming again and all four spun up but not all at once. This is odd as I’ve always had all of the motors spun up at the same time.

At first I thought I might have a power issue or some bad motors or ESC’s or something. But I found that when I connected the quad to my computer and tested the motors individually they all spun up fine. It only seemed to be a problem when I armed the quad and all four would try to start at once. I looked online but wasn’t finding any good results, not until I thought to go to BetaFPV’s site for the motors I was using. There was a link there to a support page that covered the issue I was having, which led me to looking at the startup power setting in BLHeli.

This being the first quad I’d built or used with the combination of brushless motors and 1S voltage, it had never occured to me to think about the initial amp draw that happens when the quad is armed. But it made sense once I read through the page, since the motors were fine when tested individually, and sometimes when arming I’d get a couple of motors to spin up initially and then another one or two a moment later. I ended up increasing the startup power for all four motors to 1.5 going in 0.25 increments until I could consistently get all four motors to spin up simultaneously when arming the quad.

Altogether it took me about 28 minutes to do the initial soldering, another 11 minutes to fix that solder bridge, and nearly 50 minutes to do the assembly including having to redo the motors and props and figure out the motor power issue. So the whole process was about 90 minutes, which isn’t too bad.

I’ve flown a few dozen batteries on the whoop since then and it’s quite a bit of fun to fly inside. It’s just on the edge of being too much power for indoors at least in my home which doesn’t have a lot of open space to fly. I think when I take it to the local fly-in office space it will be a lot more fun than my old brushed quad.

I also tried it outside just recently when there was some good weather and it did okay, but it’s not nearly as much fun as my toothpick is, and I only did two batteries outside out of curiosity. I think most of the time that I can fly outside it’ll be my bigger quads.

Brushless Whoop Build – Part One

It took me a while to actually sit down and do the build for the brushless whoop parts that I got for Christmas.

The first thing I like to do on a new build is test the flight controller. Especially before any soldering is done, it’s good to make sure that the flight controller works, so that later on if it doesn’t I know that it’s because of something that I did and can troubleshoot it instead of wondering if I just got a bad one from the factory. It’s easy enough to do. Plug it into a USB cable to make sure it gets power, and then make sure that Betaflight will connect to it. Test out the gyro by turning and tilting the board and make sure that the movements match up with the animated diagram in Betaflight.

At that stage, it’s not critical to do, but since it’s already connected, I like to download the latest firmware and flash it to the board. It’s just a couple of steps to do, and isn’t any harder or easier to do now or wait until later. But since it’s already connected I figure why not.

The last step I took care of before really starting the build was binding my radio. I don’t always do this part so early, usually because for anything bigger than a whoop build, I don’t use a flight controller with a built-in receiver. So there’s some soldering that normally has to take place first. But in this case since the receiver is built into the board, it’s much easier to do now where the bind button is easier to get to, than to wait and do later.

Binding the radio took me a bit to figure out since I had to find the built-in receiver specifications. I couldn’t find anything about telemetry being available or not, but it would only successfully bind when I picked the radio setting without telemetry, so I’m guessing it’s not supported. Not a big deal for a small whoop like this, but something I’ll have to keep in mind if I ever decide I need to go hunting for it outside.

With the flight controller testing out of the way, I was ready to start figuring out the build. This involves dry fitting the flight controller to the whoop frame and trying to get an idea for where the canopy, camera, and VTX were all going to sit. I found the wiring diagrams for the flight controller on BetaFPV’s site, and used those along with the printed manual that came with the VTX to figure out how everything needed to be wired together.

The only part that wasn’t immediately obvious to me until I’d studied it for a couple of minutes was that the signal wire of the camera needed to go to the camera pad on the flight controller. This would apply the OSD information, and then a wire from the VTX pad to the video in pad on the VTX would take the signal for broadcast. I also didn’t need one pair of the 5V and ground wires on the VTX since the camera already had its own.

Once I figured all of that out, then I rechecked it by fitting all of the pieces together to make sure I knew which wires to cut, which to desoldering, and what was getting soldered to what. I could always fix any mistakes, replacing or lengthening any wires that I cut too short, but I didn’t want to have to mess with that if I could avoid it with a little extra care up front.

After checking a few times though, it’s usually time to just commit and do it and hope I’ve thought through everything I needed to. In this instance that included trimming all of the VTX wires to be just a bit longer than the camera wire lengths. Trimming the plug from the camera, which I wouldn’t need, and then desoldering the two wires from the VTX that I wasn’t going to need.

Desoldering is quite easy. It just requires me to get the piece set up where I can get some pressure on the wire that I’m trying to remove, so that I can have a minimal amount of time for the soldering iron on the board. In this case the extra 5V and ground wires that were going to be replaced by the corresponding wires on the camera.

After getting the wires cut to length, I was ready to tin the wires and the pads on the VTX and flight controller. Tinning just means to prep exposed wires and any copper pads on PCB’s that I’m going to be using with solder. Ideally so that when it comes time to solder a wire to a pad, I don’t have to add any additional solder, I can simple touch them together with a bit of heat and I’m done.

Soldering a small build like this is a bit of a challenge just because of the sizes of the pads involved. I’ve found using the lowest heat setting along with a lighted magnifying glass, and a very small soldering iron tip helps tremendously for this. I’ve also applied some tricks I picked up in painting miniatures for keeping my hands steady by moving the piece to the iron and keeping my arms are my sides.

Once everything has been tinned and then soldered together, the easy part is over. With the camera, VTX, and flight controller all wired up, it’s time to check for shorts and then plugin in a battery.

Before plugging in a battery and risking releasing the magic smoke, I like to use a multimeter and check that there’s no continuity between the various connections. For example the yellow wire from the VTX to the flight controller should have continuity between the ends, but if I touch a probe to one end and then touch the other probe to the ground then I shouldn’t have any connection. If I do it means there’s a short somewhere, and on a tiny build like this it’s probably because two pads are bridged.

In my case I did have a solder bridge on the VTX between the video signal and the ground. But that’s just a matter of adding some flux to the board and using a soldering iron to try to clean up the bridge. It only took me a couple of minutes to get it cleaned up in this case, and afterwards all of the connections passed the multimeter test.

Which brings me to the last part before assembly and that was plugging in a battery and testing that everything booted up, that the VTX was broadcasting, and best of all that I was getting an image through the goggles from the camera. It’s good to do all of that before assembling everything just to make sure that there’s no more soldering fixes that need to be made, and it’s a good chance to make sure I know which way the camera needs to go so that up is up.